Powered by Blogger.

Web Analytics TV #17

Wednesday, 30 March 2011










Here are the links to the topics we discuss:




The New Google Analytics: Quick insights with Plot Rows

This is part of our series of posts highlighting the new Google Analytics. The new version of Google Analytics is currently available in beta to a small number of Analytics users. We’ll be giving access to more users soon. Sign up for early access.

The graph on top of most Google Analytics reports is designed to give you a quick overview of your site’s performance over time. From the graph it’s easy to spot trends and understand how your traffic has changed over time. One request we heard was the ability to quickly focus the graph on a particular row of data. While you could do this with a drill-down report or using an advanced segment, we saw this as an opportunity to provide an easy way to do quick comparisons in the new Google Analytics.

Say for example you’re examining your site’s traffic by traffic source. You can see there are peaks and valleys in the traffic, but if you want a sense of the major contributors, you need to dig into the table.


With Plot Rows, you can graph any two rows alongside the overview. You can then easily determine how much a row contributes to the whole. Or you can compare two lines against each other to look for comparison trends.


To use Plot Rows, just tick any one or two checkboxes next to the rows you want to plot, then at the bottom of the table, hits the Plot Rows button.


Remember, that some reports like New vs. Returning default to a Pie Chart view. This doesn’t mean you can’t use Plot Rows, just switch the view to Data, and you’re good to go.

Here’s a quick video showing this in action:


Usage Tips
When looking at continuous metrics, like Visits, Plot Rows is most revealing when exploring the rows of similar scale, for example to see how they contribute to the whole and change over time. When looking at rows at different scales the graph will be more informative when using percentage metrics like Bounce Rate.

In this example, we’re looking at organic search traffic driven to the Google Store from Google and Bing. One would not expect that Bing users are actively looking to buy Google merchandise (like this awesome t-shirt), so the number of visits is understandably low. Since the traffic from Bing is relatively low, the graph doesn’t share much we didn’t already know from the table.


In new version of Analytics you can quickly graph any of the metrics in the scorecard (the bar on top of the graph) by clicking on the metric in the scorecard. Looking at Bounce Rate, we can see that over time the Bounce Rate from Google search (orange) has dropped, which has reduced the overall Bounce Rate of the site (blue), while the Bounce Rate from Bing (green) has more or less stayed constant.

You can use Plot Rows in just about any report that has a data table. Let us know if you find a place you want this functionality that doesn’t already have it. Also, we’re planning to give a bunch more of you access to the new version this week. Be on the look out!

Accessing More Than 10,000 Rows of Data Using The API

Monday, 28 March 2011

Did you know that you can easily get more than 10,000 rows of data from Analytics using the API?
Auto-Pagination through the API

Here is a quick overview on how to do auto-pagination with our API. You might also follow along by checking the fully working sample code in Python which you can use in your own applications.

The following sample query fetches the first 10,000 keywords by conversion rate for the month of February:

https://www.google.com/analytics/feeds/data
?ids=ga:12345
&dimensions=ga:keywords
&metrics=ga:conversionRateAll
&sort=-ga:conversionRateAll
&start-date=2011-02-01
&end-date=2011-02-28
&start-index=1
&max-results=10000


Notice how start-index is set to 1 and max-results is set to 10,000. When this query is issued to the API, the API will return up to 10,000 results. The API also returns the number rows found in Google Analytics in the openSearch:totalResult XML element

14,654

To get the total number of pages in this request, we can use the following python code:

num_pages = math.ceil(total_results / 10000)

Then getting the start-index for each additional page is trivial:

for page_number in range(num_pages)[1:]: # skips the first page.
start_index_for_page = page_number * 10000 + 1


Thats it! If you want to start doing this today, or just see how it should work, we’ve included a fully working example. If you liked this, and want to see more example, let us know what we should do next in our comments.

The New Google Analytics: Dashboards

Wednesday, 23 March 2011

This is the first in a series of posts highlighting the new Google Analytics. The new version of Google Analytics is currently available in beta to a small number of Analytics users. We’ll be giving access to more users soon. Sign up for early access.

A fond farewell to Connection Speed

Issue with AdSense reports

Monday, 21 March 2011

Looking towards the future of Google Analytics

Thursday, 17 March 2011

Today at the Google Analytics User Conference in San Francisco, we shared a look at a new version of Google Analytics. We’ve also reached out to a small group of Analytics users to participate in the testing. If you’re part of this group you’ll see a link to the new version in your Analytics account. We’re starting small, and we’ll gradually roll the new version out to everyone.

Our goals for the new version are to make it easier and faster to get to the data you want and to enhance the Google Analytics platform to bring you major new functionality. Many of the changes in the new version are the result of your feedback. For example, you can now view multiple advanced segments without needing to also use All Visits. You’ll find some of the other most requested features like multiple dashboards in the new version as well.

We’ll be sharing many more details about what’s new along the way with a new series on the blog focused exclusively on the new Analytics. In the meantime, if you want to be considered for the test, visit the beta sign-up page. Using the new version won’t change how Google Analytics reports your website traffic, and you can continue to use the current version while testing the new Analytics.

As you start using the new Google Analytics, you should also check out the updated Google Analytics Help Center and the Report Finder, which will show you where to find your favorite reports in the new version. We’ve also set up a new category in the Help Forum where you can ask questions and discuss the new version.

This release wouldn’t have been possible without your feedback. Please continue to send us your feedback as you start using the new interface.

Happy analyzing!
UPDATE: 3/17/11 1:40pm PST, we fixed an issue that prevented us from collecting sign-ups. If you signed-up before 1:40pm PST, please sign-up again: beta sign-up page.

Google's Data Viz Challenge

Tuesday, 8 March 2011

If you're reading this blog, chances are very high that you value data, and chances are also high that you appreciate the presentation of data. Going beyond a spreadsheet of rows and columns and creating a visualization is not only pleasing to the eye, but it can also help put meaning and context to the data and help surface important meanings. As practical as rows and columns of numbers are, sometimes a lot more can be divined a lot faster from a simple pie chart. Or a bar graph. Or an animated motion chart :-)

Data visualization is becoming an art and a discipline. There are blogs and books devoted to it. And we're obviously believers in visualizing data, and so is Google. That's why we're giving a shout out to another Google initiative, the Data Viz Challenge.

WhatWePayFor has made their data available via an API to contestants who want to take a crack at visualizing this data, and Google is offering a $5,000 prize to the best visualization, as decided on by a panel of judges, as well as other prizes for cool entries. Deadline for entry is March 27, so get cracking - we want a Google Analytics data head to win this thing. :-)

Evolution of Analytics Benchmarking Report

Friday, 4 March 2011

This week, we are beginning a new way of providing benchmarking data to Google Analytics users. For almost three years, Google Analytics has provided a Benchmarking report for users who opt in for anonymous data sharing. In a few days, however, we will be removing the Benchmarking report from the Google Analytics interface, and replacing it with an expanded report that will be emailed directly to you.



Client Side Authorization in the API
















  if self.auth_token:
    return self.auth_token


  self.auth_token =
    self.auth_routine_util.LoadAuthToken(self.token_obj_name)


  if not self.auth_token:
    self.auth_token = self.RequestAuthToken()


  self.auth_routine_util.SaveAuthToken(self.auth_token)
  return self.auth_token









  url = ('%s?xoauth_displayname=%s' %
      (gdata.gauth.REQUEST_TOKEN_URL, self.my_client.source))


  request_token = self.my_client.GetOAuthToken(

      next='oob',
      consumer_key='anonymous',
      consumer_secret='anonymous',
      url=url)


  verify_url = request_token.generate_authorization_url(
      google_apps_domain='default')


  print 'Please log in and/or grant access at: %s\n' % verify_url
  webbrowser.open(str(verify_url))


  request_token.verifier =
'Please enter the verification code '
                'on the success page: ')


  try:
    return self.my_client.GetAccessToken(request_token)


  except gdata.client.RequestError, err:
    raise AuthError(msg='Error upgrading token: %s' % err)






  my_client.auth_token = my_auth.GetAuthToken()


except auth.AuthError, error:
  print error.msg
  sys.exit(1)




  feed = my_client.GetDataFeed(data_query)


except gdata.client.Unauthorized, error:
  print '%s\nDeleting token file.' % error
  my_auth.DeleteTokenFile()
  sys.exit(1)







Don’t miss the next Google Analytics User Conference in San Francisco

Tuesday, 1 March 2011


 

Most Reading

My Blog List